Expired70% confidenceBenchmarkExact time
In the DeepSWE 1.1 benchmark, Muse Wonder Full achieved 79.0%, exceeding Claude Opus 5.5's 74.2% and GPT-6 Astra's 74.1%.
4
Sources
70%
Confidence
Short-term (~14 days)
Relevance
10/13/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/27/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedIn the MemoryBench test, Muse Wonder achieved 85.5% accuracy, exceeding Claude Opus 5.5's 81.5% and GPT-6 Astra's 81.4%.79% similarVerifiedOn the DeepSuite benchmark, Grok 4.7 scored 71%, GPT 5.6 Sol scored 72.7%, Fable 5.1 scored 70%, and Astra Max topped the leaderboard at 74.1%.77% similarUnverifiedOn general knowledge evaluations, Muse Wonder scores noticeably lower than Claude Opus 5.5 and GPT-6 Astra, but its hallucination rate is very low.75% similarPartially VerifiedMeta's chart shows Muse Wonder beating Claude Opus 5.5 on long-horizon coding tasks: DeepSWE score 78.8 versus 74.2, SWE-Marathon score 53.1 versus 46.3.75% similarUnverifiedOn the DeepSWE v1.1 (coding) benchmark, Gemini 4 Argon scores 77.9%, Claude Opus 5.5 scores 74.2%, and GPT-6 Astra scores 74.1%.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1020382API
curl https://kongchang.com/api/v1/knowledge/claims/1020382MCP
get_claim(id=1020382)